We noticed this one a couple of weeks ago and I wrote it down. This is all the notes I have.

When recalculating an event, it wasn't recalculating the PAX times using the most up to date PAX factors. We noticed that one of our "custom" classes had the wrong PAX factor, so we updated the pax factor in our season settings. I then recalculated an event, and it didn't do anything to the results. I tried updating it a few times (using pax of 1.000 to make it easy to see the change) and then recalculated, and still nothing. I even tried to move the drivers in question out of that class, then recalculate, then move them back into that class. I believe the only way I got it to work was to create a whole new driver and number, then I re-assigned all the runs to the new driver (with the same name), then recalculate and it worked.

I know this sounds like a strange situation, why would we ever update pax factors during a season? I think it's part of a larger problem. I don't think it's recalculating properly whenever "recalculate" is pressed. This is needed because sometimes a driver is classed incorrectly to start an event so we need to update him/her partway through or at the end of the day and recalculate and I don't believe it's working.

Sorry I don't have the timing laptop so all I can point to is our results online. There are 2 problems that still exist. Our results from Event 1 are here:
http://live.axti.me/results/byIfNDB1ll

Jesse Murray (car 23) ran 62.443, and after pax shows 51.702. That appears to be a pax factor of 0.828 but our season is setup to have 0.870 for that class.

Also, Abbie Done (car 123) has raw=64.079+2 (which is 68.079) and paxtime=56.369, calculating at 0.828 again.

We had a third problem but we fixed it as I described above.

I believe we double and triple checked the SM pax factor in our season and it's definitely not 0.828. Not sure if you have our season settings or if you need me to upload them to confirm. Let me know.

It is most likely working and by design. Changing indexes in your season doesn't automatically retroactively apply that to past participants in past events.

That said, there also isn't a way to easily apply index changes to all applicable participants in all applicable events in the past. Currently, you will need to edit each affected participant and resave it to grab the new index factors. Then recalc the event.

For a variety of reasons, AXti.me is largely a few denormalized datasets and when a record (e.g. a run) is made that by itself has all relevant information without requiring reference or lookup values. It is meant to be a moment in time and retain that snapshot despite supporting lookup data changes. We need to force change upon a run instead of blindly updating.

@Chris said:
Currently, you will need to edit each affected participant and resave it to grab the new index factors. Then recalc the event.

That's what we're doing now. We were able to change the driver's class, and recalc and update the website. But the pax calc is still using the wrong factor.

Here is the link to our new results from Event #1. We recalculated the event and re-uploaded it. Jesse Murray is now in the proper SM-S class, but his pax time is wrong, still showing 51.702. We also changed Abbie Done from SM to AM (pax of 1.000) just to see what would happen and again his class label changed, but not the calculated pax time.

@canadabeau I'll take a look

@canadabeau I did my own test and it worked as intended.

1. Changed a class's index in Manage Seasons
2. Open an event
3. Selected a driver that was in the class I changed
4. Opened the edit registration for the driver and checked the Apply to Existing Runs and clicked Save
5. Went to event home and clicked recalculate event
6. Opened the Results screen and see that the driver's pax standing changed and paxed time changed as necessary.

For the calculation of pax time, the index is used from the class data stored on the run.